Evidence-based approaches to creating gender-equal organisations
At the Innovation Growth Lab, we have been working with our policy and research partners to support women in innovation and entrepreneurship for years. This International Women’s Day, we are flipping the script: instead of asking how business departments and innovation agencies can create programmes and policies that serve female inventors and entrepreneurs better, we want to explore how public sector organisations themselves can become more inclusive employers.
Despite the public sector doing better than the private sector in terms of gender equality, women still earn less than men on average, and they are underrepresented within top civil sector jobs. Unconscious bias training has become a popular diversity initiative for organisations attempting to remedy discrimination. However, a new study suggests that unconscious bias training alone is unlikely to solve the challenge. To inspire organisations to go beyond training, we review five evidence-based approaches that promote equal pay and fair progression opportunities for women in the workplace.
Organisations may find it tempting to start addressing their lack of diversity by adding a ‘token’ woman to an all-male team. Yet, findings suggest that if organisations want to unlock the benefits of diversity for group decision making, they need to create gender-balanced teams: adding a token woman will not suffice.
A recent experiment that involved randomly assigning university students to semester-long teams provides a cautionary tale against token inclusion of women. Results from the study show that lone women in majority-male teams have a harder time making their voice heard and influencing group decisions than women on majority-women teams. In their setting, token women are less likely to be chosen to represent the group, participate slightly less in group discussions and receive less credit when they do.
Fair task assignment
An often overlooked determinant of career progression is task allocation: whether someone is asked to perform challenging or routine tasks in their roles affects their ability to hone their skills, learn on the job and demonstrate their motivation. There are two aspects of job assignment to consider: who is offered the challenging tasks, and who is asked to take care of the “office housework” that will never lead to a promotion. Women are disadvantaged on both counts: they are less likely than men to be assigned challenging tasks at work, and they tend to volunteer, are asked to volunteer, and accept requests to volunteer more often than men for tasks that everyone prefers be completed by someone else.
The authors of the above experiment demonstrating gender differences in low-promotability tasks provide practical recommendations for organisations, including tracking task assignments to recognise inequalities, rotating assignments (using a lottery or a predefined schedule) rather than asking for volunteers, and re-bundling tasks within jobs, moving tasks that are non-promotable for one job to other jobs for which the task would be profitable.
Performance evaluations are ubiquitous, consequential – and often biased. Organisations can make their evaluation processes fairer by tracking whether there are systematic gender differences in their appraisals and educating their evaluators about the existence of bias.
Bias in evaluations is often hard to prove because it is rarely possible to measure people’s “true” performance. To overcome the challenge of identifying bias, a researcher analysed student evaluations at a French university, a setting where objective measures of teaching quality (instructor’s ability to make their students succeed on their final exams) are available. She found that female instructors received systematically lower evaluation scores (particularly from male students), despite not doing any worse than male instructors according to the objective performance measures. In a follow-up experiment, the author of the original study on student evaluations teamed up with a colleague to test a simple and effective way to improve evaluations. They demonstrated the success of an awareness raising campaign (that shared precise information with students about past discrimination) in reducing bias against female instructors.
A large body of literature shows that negotiation tends to disadvantage women and can increase the gender pay gap. Consider the example of teachers in Wisconsin: until 2011, male and female teachers with the same credentials earned the same on average – but since then, female teachers’ salaries have declined relative to their male colleagues. What happened? A recent study links the emergence of this gender pay gap to a legislative change that introduced individual bargaining and flexible pay for teachers. Surveys suggest that female teachers felt much less comfortable negotiating for higher wages (especially when their superintendent was a man), and as a result, secured lower wages than their male colleagues who did negotiate.
Organisations should be wary of initiatives that try to ‘fix’ women by encouraging them to negotiate more. “Women do not appear to be broken” – it is the institutions that need to change. Transparency can help: making it clear what is negotiable, and what to expect from negotiations may eliminate the gender gap in negotiation outcomes.
There is robust evidence from various countries quantifying motherhood’s toll on women’s earnings both in the short and the long run. In addition, the pandemic has highlighted and worsened gender inequalities in housework and caring responsibilities, with the increased burden of unpaid work largely falling on women. Flexible work arrangements are key to helping working mothers, and encouraging employers to offer them is a promising path forward, but such arrangements may themselves exacerbate gender progression gaps if managers are biased against remote workers.
Organisations committed to gender equality should thus actively disrupt gendered expectations around care work by offering generous parental leave policies for all new parents, irrespective of their gender or their level of biological involvement, and testing interventions to change norms around the actual uptake of parental leave.
Generating more evidence
Going forward, we hope to see more public sector organisations apply these insights in their own personnel policies – ideally doing so in an experimental fashion, collecting data to test the effectiveness of the approaches in their own contexts.
It is also important to recognise the limitations of the evidence base: much of the experimental literature we have covered applies a narrow gender-binary framework and lacks intersectional considerations. As such, there is a lot of room for improving our understanding of what works and for whom – and our team is keen to collaborate with organisations who aspire to create a fairer workplace. Find out how we can work together!